There's one widespread reason for optimism which I think is worth questioning. It's the idea that a fall in inflation next year will help to reverse the fall in real wages and so raise real households' incomes.

First, let's check the data. Since 1949 - when current records began - the correlation between annual inflation and annual growth in real personal disposable incomes has been minus 0.31. This seems to support the optimists' case; lower inflation is associated with higher real incomes.

However, much of this correlation is due to the mid-70s period, when a soaring oil price cut the real incomes of oil-consuming nations - a process exacerbated by the government's attempt to use wage control to fight inflation. Since 1980, the correlation has been a statistically insignificant minus 0.15.

There's a good reason why the correlation should be insignificant. It's that inflation has ambivalent effects upon real incomes.

If inflation falls because of a positive supply shock - such as a drop in commodity prices - then we would expect to see real wages and incomes rise.

However, if inflation falls because of weaker aggregate demand, then labour demand will be weak, which will hold down real incomes. Between the late 80s and the early 90s, for example, inflation fell but so too did income growth, precisely because of that weak demand.

Which brings me to our current predicament. There's no good reason to suppose that commodity prices will fall except because of random noise; remember Hotelling's rule? We cannot therefore bet on this source of a negative correlation between inflation and real income growth asserting itself. Granted, we could get the positive supply shock of a return to productivity growth, but this is more likely to benefit capitalists than workers in the first instance.

However, it's quite possible that inflation will fall because of weak aggregate demand and excess capacity. But these are conditions in which real wages won't rise much and might even continue to fall.

In saying all this, I'm not ruling out a recovery in real incomes next year; I know nothing about the future. It's just that if you want to contend that this will happen, you should simply contend that aggregate demand - and with it workers' bargaining power - will pick up briskly. Leave inflation out of it.